Planetary Resources

[FutureSpaceTourist]

From the first KS update video:

the [Arkyd 100] field of view is about a degree and a half

[neilh]

The first project update to Kickstarter supporters from Planetary Resources mentions that Peter Diamandis, as well as Lori Garver, former astronaut Ed Lu, and Bill Nye, are participating in a White House "#WeTheGeeks" Google+ hangout at 2pm ET Friday (March 31):

https://plus.google.com/+whitehouse/posts/PiQiTv4bbs3
https://plus.google.com/events/cv333b50h80jvrejdmq77p49gsc
http://www.whitehouse.gov/webform/lets-talk-about-asteroids

On Friday, May 31st at 2pm EDT, the White House will host a "We The Geeks" Google+ Hangout on asteroids. During a conversation moderated by OSTP's Cristin Dorgelo, we’ll be talking about asteroid identification, characterization, resource utilization, and hazard mitigation ...

[guckyfan]

From the first KS update video:

the [Arkyd 100] field of view is about a degree and a half

Just to give it some perspective, the moon has 1/2 degree diameter, so it is three times the moon diameter.

[Tass]

Hmmmm. Using commercial CCDs with likely aspect ratio, that could be comparable to the Olympus E1 with 2560x1920=5.1MP, or the Nikon D1x 4024x1324=5.24 MP. I figure the 1 degree FOV applies to the length and then the width is less than 1 degree. Otherwise you have to focus more, have 'faster' primary, which means more manufacturing, and they're minimizing cost. So that's one question for the FAQ - is the advertised FOV of 1 degree across the shorter width or the longer length?
http://forums.steves-digicams.com/newbie-help/31261-megapixel-ccd-size.html

And that's the first I heard the nm range. Resembles the science capability in:
http://www.skphotonics.com/pdf/la_pdf/CCD_Camera_Detector.pdf
"Each camera is calibrated in the factory after extensive thermal cycling." Ok, so no biases at the customer level. Operating temperature 0 to 40 C, so maybe it'll have to be actively heated?
I'm very glad to hear the H alpha filter. A 1 degree FOV will get all the sun, and that will be a beautiful thing. Or.... did someone say you can't point this at the sun?

Yes, they said anywhere but the sun. And it seems they have retracted on the Earth too, but I am not entirely clear on that.

[Lar]

To carefully represent myself, I just finished a incorrect Astronomy (my second bachelors)

Your ship jumping problems are because you got an "incorrect" degree... the "correct" ones are much more favored by employers.

I thought at one point they were saying A 100's could be pointed at earth... maybe they need a variant that can handle the additional brightness?

And even making 5 or 10 at a time counts as "mass production" in the space biz... to me anyway

[jebbo]

I assume you're getting radial velocity at WIYN? And what's the connection between planethunters and eclipsing binaries.

As EBs are a major source of false-positives, we're also cataloguing the ones in the Kepler FOV.  So far, we have a list of over 400 that weren't in Prsa's original catalogue (though Andrej has since incorporated almost all of them).

I don't think we're looking for RV data, more for spectrographic data on the stars as most have just Sloan "griz" data.  Though there are a few EBs with evolving orbits (from TTVs, TDVs, etc) where RV data might be useful.

There is also some work going on using mutually eclipsing triples for stellar mass determination (and some of those light curves are fabulous to behold:-) ).

FYI: I'm using "we" here in a generic sense to cover both the official science team (which is very small) and the small group of serious amateurs who can analyse candidates (of which I'm one).

I only checked it for a few minutes and it looks like crowd sourced identification of transits. I could surmise that if you have good primary and secondary transits in a binary, you ~~may~~ be able to pull out timing and transit width variations caused by non-transiting planets. Wild speculation on my part.

The primary purpose is to gather user characterisations of light curves which are then pushed through an automated pipeline to identify possible candidates.  This has been very successful for short period exoplanets and the reliability has been fairly well calibrated by also inserting simulated transits . . .

However, as well as this, there is a very successful ad hoc science pipeline run through "chat" that has evolved over time.  This has found (my guesstimate) ~150+ candidates of which about 20% have not been identified by the automated pipelines (transits on very active stars; circumbinary transits; etc).  What has been slightly irritating is that many have now been published by others (and there is a suspicion there has been some unacknowledged data mining:-( ).

However, there are two confirmed exoplanets and 42 additional candidates:
- http://arxiv.org/abs/1210.3612 (Kepler-64b aka PH1 b)
- http://arxiv.org/abs/1301.0644 (PH2 b)

--- Tony

[Robert Thompson]

Yes, they said anywhere but the sun. And it seems they have retracted on the Earth too, but I am not entirely clear on that.

Ok, this isn't confusing. These are ultimately intended to perform science on cool objects. A solar H-alpha would be too opaque for science on asteroids, ergo, no possibility of (I wager) of a solar H-alpha image without burning the CCD, and so just no. You can still photograph hydrogen with enough time, for instance 3.5 hours:
www.rc-astro.com/photo/id1065.html

I'm not immediately appreciating why you are doing spectroscopy on false positives. Are you exploiting them for science of their own merit, or for further characterization of false positives (in which case I am confused since Kepler data won't reflect spectrum)?

I hadn't heard of triple eclipsers. I googled and lo:
http://kepler.nasa.gov/news/nasakeplernews/index.cfm?FuseAction=ShowNews&NewsID=119
with a Beethovenesque:
http://kepler.nasa.gov/images/videos/PR04_2.mov
Some app writer should put lightcurves to music.

To genuflect towards topic, Arkyd could easily do something like EB or triples.

Your account of the "chat" pipeline is analogous to arguments in favor of human science space flight….

[yg1968]

Planetary Resources has reached their $1M goal for their (crowdfunding) Kickstarter project:
http://www.kickstarter.com/projects/1458134548/arkyd-a-space-telescope-for-everyone-0/posts/515328

See also this thread:
http://forum.nasaspaceflight.com/index.php?topic=32029.120

http://www.youtube.com/watch?feature=player_embedded&v=YLbkFbE-8Z4#at=20

[JohnFornaro]

From the first KS update video:

the [Arkyd 100] field of view is about a degree and a half

Just to give it some perspective, the moon has 1/2 degree diameter, so it is three times the moon diameter.

At the distance of the Moon from Earth.  By the time you get to the asteroid belt, your cone of vision has a large diameter, and the rocks are still pinpricks.

One technical issue I struggle with is how their camera can have enough resolution to look at the pinpricks; stabilize its view so as to follow multiple pinpricks in order to get valid data on their orbital characteristics; AND... accurately observe the tiny fluctuations in the brightness of the objects so as to accurately characterize their rates of tumble, and somehow note their size and mass.

[Garrett]

One technical issue I struggle with is how their camera can have enough resolution to look at the pinpricks; stabilize its view so as to follow multiple pinpricks in order to get valid data on their orbital characteristics; AND... accurately observe the tiny fluctuations in the brightness of the objects so as to accurately characterize their rates of tumble, and somehow note their size and mass.

Just randomly jumping into this conversation, and am not an astronomy/Planetary Resources expert, but here's my take:
Camera resolution and orbital characteristics: I don't think they need very high resolution for the Arkyd 100. If they can record an asteroid's "pinprick" signal at different times throughout the year, I'm pretty sure that the maths of orbital mechanics will still give them a very good idea of the asteroid's orbit.
Rates of tumble, size, mass: Is that an objective of Arkyd 100? My impression was that such data was not going to be available until Arkyd 200.

There are probably also other methods for estimating an asteroid's mass, such as from irregularities in its orbit. Hopefully an astronomer will jump here and enlighten us.

[oldAtlas_Eguy]

One technical issue I struggle with is how their camera can have enough resolution to look at the pinpricks; stabilize its view so as to follow multiple pinpricks in order to get valid data on their orbital characteristics; AND... accurately observe the tiny fluctuations in the brightness of the objects so as to accurately characterize their rates of tumble, and somehow note their size and mass.

Just randomly jumping into this conversation, and am not an astronomy/Planetary Resources expert, but here's my take:
Camera resolution and orbital characteristics: I don't think they need very high resolution for the Arkyd 100. If they can record an asteroid's "pinprick" signal at different times throughout the year, I'm pretty sure that the maths of orbital mechanics will still give them a very good idea of the asteroid's orbit.
Rates of tumble, size, mass: Is that an objective of Arkyd 100? My impression was that such data was not going to be available until Arkyd 200.

There are probably also other methods for estimating an asteroid's mass, such as from irregularities in its orbit. Hopefully an astronomer will jump here and enlighten us.

Lets see.

A 32mega pixel HD camera has over 4000 pixels in the smaller dimension. BTW a proper designed HD camera has exactly square pixels which drives the normal pixel number combination 1920x1080 HD pixel resolution. A 32MP camera has a total of 16 times the number of pixels (a 4x 4 array of pixels that make up the single HD pixel which is used for electronic stabilization and for a electronic 4 X zoom.

With 4000+ pixels to cover an arc of 1 degree thats .00025 degrees per pixel. At the distance of the Moon thats 1.7km per pixel resolution. With black to full resolution of just 1000/1 that can make the detection of objects 100 times smaller or those about 170ms in size or mabye a little smaller if the contrast ratio is even higher 12bits 4000/1 or 16 bits 64000/1.

Although this telescope can find the small asteriods it will not be able to give much details other than brightness. spectra and data leading to orbit determination.

[JohnFornaro]

One technical issue I struggle with is how their camera can have enough resolution to look at the pinpricks; stabilize its view so as to follow multiple pinpricks in order to get valid data on their orbital characteristics; AND... accurately observe the tiny fluctuations in the brightness of the objects so as to accurately characterize their rates of tumble, and somehow note their size and mass.

Just randomly jumping into this conversation....

With 4000+ pixels to cover an arc of 1 degree thats .00025 degrees per pixel. At the distance of the Moon thats 1.7km per pixel resolution. With black to full resolution of just 1000/1 that can make the detection of objects 100 times smaller or those about 170ms in size or mabye a little smaller if the contrast ratio is even higher 12bits 4000/1 or 16 bits 64000/1.

Although this telescope can find the small asteriods it will not be able to give much details other than brightness. spectra and data leading to orbit determination.

Thanks for jumping in. 

They'll not be looking in the vicinity of the Moon, but rather the vicinity of the asteroid belt.  This drastically changes the area under observation.

Their camera's got to be higher resolution than what you suppose, which would not be able to see the rock that is intended to be retrieved. 

Clearly there is an adjustable lens involved, and the FOV must be much narrower at the expected distance.  That's for a sat in LEO or GEO.  If they fly on out to the belt, there will be better resolution available, but there will be communication trade-offs, as well as a much wider field of view to examine.

[D_Dom]

With a minimal understanding of the described capability I think spectra means wavelength of the reflected light. This leads me to wonder if the surface material composition can be identified.

[jebbo]

AIUI, the primary goal of the ARKYD-100 is finding asteroids in near-earth orbits (not main belt objects), rather than characterising them.

So I think their fixed focal length f/4, 200mm telescope is adequate: basically take 3 photos and look for moving objects.  From the 3 positions of each moving object, work out the orbit.  My guess is this is done without filters for maximum light gathering capability.

Once you've found a bunch of asteroids in suitable orbits, you can follow-up with ground-based instruments to get the size and (possibly) composition but this is very dependent on getting instrument time so might only be done for a select few objects.

The telescope also has a bunch of filters: UV bandpass (< 300 nm), B, V, R, OIII, Hα, 1 μm bandpass, Luminence (Clear). 

This means you can also do space-based follow-up and compare brightness at different wavelengths to get an idea of composition, and likely asteroid class.  But this is not a full spectrograph so you don't get a composition.

Once you have an asteroid class, you can get a rough size and mass estimate from the brightness of the object.

That's my best guess, but I'm not an astronomer nor do I play one on TV

[Vultur]

They'll not be looking in the vicinity of the Moon, but rather the vicinity of the asteroid belt.  This drastically changes the area under observation.

As jebbo says, I believe they are looking for NEOs, not in the asteroid belt.

[JohnFornaro]

AIUI, the primary goal of the ARKYD-100 is finding asteroids in near-earth orbits (not main belt objects), rather than characterising them.

Understood.  But the first suggestion had to do with camera resolution in a 1.5 degreee FOV at lunar distances.  I followed up with the general characteristics at the asteroid belt.

Those NEO's are in a wide range of orbital characteristics, eccentricity, inclination, mass, etc.  I'm not considering the known objects, since they say they are specifically looking for unknown objects with unknown orbital characteristics.

Obviously, they're not looking towards the Sun, for brightness and resolution reasons.  So I'm trying to figure out their strategy.

Are they going to be looking at a specific area of the sky, basically waiting to see what crosses over their FOV?  That would certainly be one way of looking.

If they had a fleet, each one looking at a slightly different, but still fixed FOV, that would increase their chances of finding something, since there'd be more "eyes" looking.

If they only have one satellite, could they be planning on panning the sky with some known slew rate?

The resolution of their camera may be considered as a filter.  Automatically, it would only respond to NEO's that are well, in Near Earth Orbit.

They have not yet told the public sufficient information about the resolution of the camera, nor the strategy for looking, which is why, I suppose, it is being speculated about on this thread.

But it sounds like a simple strategy.  Sit and wait; point and shoot; count and calculate.  Repeat.

[jebbo]

Obviously, they're not looking towards the Sun, for brightness and resolution reasons.  So I'm trying to figure out their strategy.

Actually, I'm not convinced of this given the filter-set as some are pretty specific to looking at the sun (e.g. H-alpha).  I suspect shots of the sun will be a popular choice for those who've backed it sufficiently to get photos of their choice.

However, when they are asteroid hunting, I agree completely.  They will probably want to avoid very bright stars as well.  Assuming they are using a CCD (I can't imagine they're using a CMOS sensor!) and want long integration times so push the magnitude limit they will want to avoid charge-bleed, etc.

If they only have one satellite, could they be planning on panning the sky with some known slew rate?

I think I've read somewhere that the publicly accessible one is the 2nd they are launching (discounting Arkyd-3 which is an electronics test) but darned if I can remember where

They have not yet told the public sufficient information about the resolution of the camera, nor the strategy for looking, which is why, I suppose, it is being speculated about on this thread.

Yup.  Not just resolution but a whole gamut of important things: the per pixel point spread function of the optics, noise floors, dark levels, mapping of pixels onto sky coordinates, what their data release strategy is (will they release FITS files, etc).

I think some of this might be in the Kickstarter comments thread.  I attempt some mining there:)

But it sounds like a simple strategy.  Sit and wait; point and shoot; count and calculate.  Repeat.

Indeed.  But why be complicated?  Spotting objects moving against the background is straight-forward and largely automated (subtract 2nd shot from 1st and anything that isn't zero is moving).

[Robotbeat]

Limiting resolution isn't fundamentally constrained by the number of pixels on the sensor, but by the aperture size and then wavelength, to first order. Second order, you have signal to noise ratio, etc.

If you want to figure out their strategy, start with what the amateurs do on the ground to get something of an idea. Also, you can read papers about proposals for asteroid survey missions.

[ChefPat]

PR says they will equip their Arkyd 100 series with a cutting edge Laser Communication system. Will they be able to use multiple Arkyd's as an Interferometer utilizing the Communication system?
If they can do that they can exponentially increase their optical resolution.

[neilh]

PR says they will equip their Arkyd 100 series with a cutting edge Laser Communication system. Will they be able to use multiple Arkyd's as an Interferometer utilizing the Communication system?
If they can do that they can exponentially increase their optical resolution.

I'm not sure, but I believe the Arkyd 100 reuses the same optics assembly for both observation and laser communication, so both wouldn't be possible simultaneously on the initial model.